False alarms are a significant problem for security systems because the alarms result in a waste of resources. Specifically, a remote monitoring station receives the alarm from the control panel or sensor and commences a response. The response can include calling the local police or fire department. The police or fire department responds by traveling to the protected property and investigating the alarm. Meanwhile, a real emergency might be occurring at other locations. Additionally, there is a potential for a fine or penalty for misuse of police resources. False alarms can be generated as a result of environmental changes, human error, errors in a sensitivity setting and pets moving within a protected area.
U.S. Pat. No. 7,106,193, issued on Sep. 12, 2006 to Kovach and assigned to Honeywell International Inc., describes an alarm detection and verification device. The alarm detection device includes two sensors, a primary sensor and a secondary sensor. The verification device includes a verification sensor such as a video camera. The alarm is first detected and then verified. The detection of alarm condition is based upon a binary decision process, i.e., yes or no. In other words, the detection of an event is an all or nothing decision process.
Similarly, U.S. Pat. No. 4,857,912, issued to Everett Jr. et al., on Aug. 15, 1989, describes a multi sensor security system where the detection of alarm condition at each sensor is based upon an on or off state of the output.
However, using such a decision criterion does not account for the raw data included a sensor output or activity that is just below a detector threshold. False alarms can be generated where the sensor outputs an incorrect “on” or “off” state